Qwen-2.5-7B-Simple-RL / all_results.json
hzk17's picture
Model save
fa9d502 verified
{
"total_flos": 0.0,
"train_loss": 0.09447722357977083,
"train_runtime": 119506.5825,
"train_samples": 30000,
"train_samples_per_second": 0.251,
"train_steps_per_second": 0.008
}